Updated 30 October 2026: To get started with Claude Code, install it with Anthropic's one-line native installer, open a terminal in your project folder and type claude. The first time, your browser opens so you can sign in with a paid Claude plan (Pro, Max, Team or Enterprise) or a Claude Console account. Then describe a small task in plain English, approve the changes you are happy with, and add a CLAUDE.md file so Claude remembers how your project works.
Who this guide is for
This guide is for you if you want to start using Claude Code, Anthropic's AI coding agent, and you are new to it, new to the terminal, or following older instructions that no longer work. A coding agent is an AI tool that can read your project, edit files and run tasks for you while you describe what you want in plain English.
You do not need to be a developer. You do need to be comfortable copying a line into a terminal, which is the text window where you type instructions to your computer instead of clicking. This guide walks you through every step.
If you are still deciding between tools, read Cursor vs Claude Code for beginners: which should you learn first? first.
What you will need
- A paid Claude plan or a Claude Console account. Claude Code works with Pro, Max, Team and Enterprise plans, or with a Console account (pay as you go API access). The free Claude plan does not include Claude Code.
- A supported computer. Anthropic lists macOS 13 or later, Windows 10 (version 1809) or later, Ubuntu 20.04 or later, Debian 10 or later, or Alpine Linux 3.19 or later, with at least 4 GB of memory.
- An internet connection.
- A project folder. An existing project or a brand new, empty folder both work.
Step 1: Open a terminal
- Mac: press Cmd and Space, type Terminal and press Return.
- Windows: right-click the Start button (or press Windows and X) and choose Terminal or PowerShell. When you are in PowerShell, each line starts with "PS". In the older Command Prompt (CMD), it does not.
- Linux: press Ctrl, Alt and T on most distributions.
If you use Cursor or another code editor, you can open its built-in terminal instead. That way Claude Code runs right next to your files.
Step 2: Install Claude Code
Anthropic now recommends its native installer. Older guides, including the first version of this post, used npm (a package manager for JavaScript tools). That route still exists, but it is no longer the recommended one, and you do not need Node.js for the native installer.
Copy the line for your computer, paste it into the terminal and press Return (Enter on Windows). The installer shows no progress bar while it downloads, so give it a moment.
Mac, Linux or WSL (Windows Subsystem for Linux):
curl -fsSL https://claude.ai/install.sh | bash
Windows PowerShell:
irm https://claude.ai/install.ps1 | iex
Windows Command Prompt (CMD):
curl -fsSL https://claude.ai/install.cmd -o install.cmd && install.cmd && del install.cmd
You do not need to run these as an administrator. On native Windows, Anthropic recommends also installing Git for Windows. Without it, Claude Code uses PowerShell to run commands instead.
Native installs update themselves in the background, so you stay on the latest version. Anthropic also offers Homebrew on Mac and WinGet on Windows, but those do not update automatically.
Step 3: Check it worked
Close the terminal, open a new one and type:
claude --version
A version number means you are ready. If you see "command not found" (Mac and Linux) or "not recognized" (Windows), your terminal has not picked up the install yet. Close every terminal window, open a fresh one and try again. If it still fails, Anthropic's setup page has a troubleshooting section for exactly this.
Two Windows mix-ups are common. If you see an error about "&&" not being a valid statement separator, you are in PowerShell, so use the PowerShell line. If you see "irm is not recognized", you are in CMD, so use the CMD line.
Step 4: Start Claude Code in your project
Move into your project folder. The simplest way is to type cd followed by a space, drag your project folder from Finder or File Explorer into the terminal window, then press Return. If you opened the terminal inside your code editor, you are usually in the right place already.
Then type:
claude
Claude Code opens an interactive session in your terminal. It works on the folder you start it in, so always start it inside your project, never in your whole home folder.
Step 5: Sign in
The first time you start Claude Code, it asks you to log in and opens your browser. You can sign in with:
- A Claude subscription (Pro, Max, Team or Enterprise). Anthropic recommends this for most people.
- A Claude Console account, which uses pre-paid API credits. On first login, Claude Code creates a "Claude Code" workspace in your Console so you can track costs.
Teams can also connect Claude Code through cloud providers such as Amazon Bedrock, Google Cloud or Microsoft Foundry, but you will not need that to get started.
Once you are signed in, Claude Code remembers you. To switch accounts later, type /login inside a session.
Step 6: Give it your first task
You are now in a conversation with Claude Code. Start by asking questions, then move on to a small change:
- "What does this project do, and how are the folders organised?"
- "Explain the main file in simple terms."
- "Add a simple contact page with a name, email and message field."
For each change, Claude Code shows you what it wants to do. If it asks before making the change, read it and approve it if you are happy. Then check the result yourself, for example by refreshing your app in the browser.
Small, specific requests work far better than vague ones. "Build me an app" goes badly. "Add a sign-up page that asks for name, email and password" goes well.
Step 7: Add a CLAUDE.md file
CLAUDE.md is a plain text file in your project that Claude Code reads at the start of every session. Think of it as a briefing note: what the project is, how to run it, which tools you use and any rules to follow.
The quickest way to create one is to type /init inside Claude Code. It looks through your project and writes a starting CLAUDE.md for you. If you already have one, it suggests improvements instead of overwriting it. Then add the things Claude could not work out on its own, such as "always use British English in the interface".
For a deeper dive, read The ultimate guide to CLAUDE.md. If you also use Cursor or Codex, see One AGENTS.md for Claude Code, Codex and Cursor so all your agents follow the same rules.
Everyday basics
| What you want to do | How |
|---|---|
| See the available commands | Type /help |
| Pick up a previous conversation | Type /resume |
| Stop Claude mid-task | Press Esc |
| Change how much Claude does without asking | Press Shift and Tab to switch permission mode |
| Switch account or sign in again | Type /login |
| Leave Claude Code | Type /exit, or press Ctrl and D twice |
A few habits that make a big difference:
- Save your work often. Commit to Git (save a snapshot of your project) every time something works, so you can always go back.
- Ask for a plan first on bigger jobs. Ask Claude to explain how it will approach a task before it changes anything.
- Read before you approve. Claude Code asks for permission for a reason.
- Teach it repeatable jobs. Once you are comfortable, skills, hooks and plugins let you give Claude reusable instructions and automatic checks. Our guide to Claude Code skills, hooks, plugins and mods explains each one.
Other ways to use Claude Code
The terminal is the original and most flexible way in, but it is not the only one. Anthropic also offers a desktop app with a Code tab, extensions for VS Code and JetBrains editors, and a web version at claude.ai/code. If you prefer clicking to typing commands, see How to get started with coding in Claude Desktop.
Common mistakes
- Following old npm instructions. Use the native installer above instead.
- Using the free Claude plan. It does not include Claude Code, so sign-in will not give you access.
- Starting Claude Code in the wrong folder. Start it inside your project so it only sees what it needs.
- Reusing an old terminal window after installing. Open a new one so it can find Claude Code.
- Huge, vague requests. Break work into small steps and check each one.
- Skipping CLAUDE.md. Without it, you will repeat the same instructions in every session.
FAQs
How do I start Claude Code in the terminal? Open a terminal, move into your project folder and type claude, then press Return. The first time, you will be asked to sign in through your browser.
How do I install Claude Code? Use Anthropic's native installer: the curl line on Mac, Linux or WSL, or the PowerShell or CMD line on Windows, all shown in Step 2. Then check it worked by asking for the version number.
How do I open Claude Code after installing it? Every time you want to use it, open a terminal in your project folder and type claude. You only need to install and sign in once.
How do I use Claude Code in the terminal once it is running? Type what you want in plain English and press Return. Claude Code reads your project, proposes changes and asks for approval where needed. Type /help to see the available commands.
Is Claude Code free? No. It needs a paid Claude plan (Pro, Max, Team or Enterprise) or a Claude Console account with API credits.
Do I need Node.js or npm to install Claude Code? No. The native installer does not need Node.js. The npm route still exists, but Anthropic recommends the native installer.
Can I use Claude Code inside Cursor? Yes. Open Cursor's built-in terminal and type claude. Many builders work this way, and our Claude Code course includes a lesson on it.
Learn Claude Code properly
The fastest way to get confident is to build something real. In Build a full stack Next.js app with Claude Code, you install and configure Claude Code, learn to use it inside Cursor, and build a full stack project management app with Supabase, Stripe, Resend and the AI SDK, then deploy it for real customers. No prior coding experience is needed.